In the logistics operation of unmanned storage, there will be mixed goods with both simple and complex demand types. To ensure the effective distribution of these mixed goods, the hybrid task assignment problem of a heterogeneous robot system is worth studying. A novel algorithm based on the Fuzzy C-Means algorithm called the Fuzzy C-means-consensus-based bundle algorithm (FC-CBBA) is proposed. This algorithm incorporates complex constraints and uses the consensual-based bundle algorithm (CBBA). The logistics operations with different demand types in the storage are defined as hybrid tasks, and the execution priority of complex tasks is set by the time window. Firstly, the FCM algorithm is used to divide the position of the task into soft clustering regions according to the membership degree. Then, the robot grouping resource allocation strategy is proposed to minimize the number of robots in the region. Finally, the FC-CBBA algorithm is used to allocate tasks for each region. The performance of the algorithm is verified in MATLAB, and the results show that the proposed FC-CBBA algorithm can shorten the allocation time and achieve a task completion degree of 96.21%. Compared with CBBA and the consensus-based bundle algorithm (CBCA), the proposed algorithm shows advantages in timeliness, and task completion, which reflect the effectiveness of the proposed algorithm in solving the hybrid task assignment problem.